Imaging speed is a crucial consideration for numerous clinical and research applications of MRE. In recent years, parallel magnetic resonance imaging (PMRI) has been shown to be an effective means of increasing MR imaging speed beyond previous limits. In contrast to traditional sequential MR acquisitions, which encode image data one point at a time, PMRI techniques use arrays of RF coils to encode and detect data in a parallel rather than a strictly sequential fashion. PMRI techniques have been advancing rapidly, and numerous potential applications have been identified. However, certain basic questions remain regarding the image quality which may routinely be achieved in PMRI scans. Two particular challenges relate to the areas of coil sensitivity calibration and signal-to-noise ratio optimization. We propose a program of research to address each of these basic challenges, by eliminating the need for separate sensitivity calibration scans, and by designing coil arrays specifically tailored for spatial encoding. The practical clinical value of these technical developments will be verified in a targeted clinical study of patients with known or suspected abdominal pathology. Specific Aims are as follows: 1. Optimization and comparative evaluation of a new generalized self-calibrating parallel MIRI technique Optimization of known free parameters in the self-calibrating approach Comparison with standard externally-calibrated PMRI approaches in phantoms, volunteers, and patients 2. Construction and comparative evaluation of novel RF coil arrays explicitly designed for parallel MIRI Design and construction of RF coil arrays tailored for spatial encoding, including arrays with independently positionable elements, 'multipole" arrays, and arrays of arrays ("superarrays') Comparative evaluation in phantoms, volunteers, and patients Exploration of the computational tractability of ultimate intrinsic SNR calculations for PMRI studies with arbitrary arrays 3. Implementation and evaluation of the developments of Specific Aims 1 and 2 in a clinical study of abdominal imaging using parallel MIRI Use of an externally-calibrated SMASH technique to reduce breath-hold durations and/or increase slice coverage in a clinical liver imaging protocol using an existing coil array. Comparison of clinical images obtained using the optimized self-calibrating technique from Specific Aim #1 with images obtained using externally-calibrated techniques in the same coil array Comparison of clinical images obtained using the coil arrays from Specific Aim #2 with images obtained using an existing standard array